Same income, same year, different fit
Spain and Czechia produce almost exactly the same income per person, 59,868 dollars against 60,487, a difference of one per cent. In 2024, 30.1% of Spain’s employed held more education than their occupation requires, against 10.8% of Czechia’s: 7th and 79th of the 107 economies in the set, and a ratio of 2.8.
The obvious reading, that Czechia simply matches its workers better, only half survives the rest of the row. Czechia does match more of them, 66.8% against Spain’s 60.5. But its remaining mismatch points the other way: 22.3% of Czech workers hold LESS education than their job calls for, against 9.5% in Spain. Spain leaves two workers in five mismatched either way, Czechia one in three.
The spread across the set is wide at both ends. South Korea is the most over-educated of the 107 at 40.9%, and Niger the least at 0.8%, where 90.7% of the employed hold less education than their occupation requires. The median is 17.2%, which is Guyana’s own figure; 7 economies are above 30% and 24 below 10%.
| Economy | Year | Over-educated | Matched | Under-educated | Rank of 107 |
|---|---|---|---|---|---|
| South Korea | 2025 | 40.9% | 46.9% | 12.2% | 1 |
| Spain | 2024 | 30.1% | 60.5% | 9.5% | 7 |
| Philippines | 2023 | 29.6% | 43.1% | 27.3% | 8 |
| United States | 2025 | 29.5% | 54.2% | 16.3% | 9 |
| United Kingdom | 2025 | 26.4% | 53.4% | 20.3% | 14 |
| Türkiye | 2025 | 22.4% | 51.3% | 26.3% | 28 |
| France | 2024 | 22.4% | 60.2% | 17.4% | 29 |
| Poland | 2024 | 20.0% | 68.5% | 11.6% | 43 |
| Vietnam | 2024 | 19.8% | 58.6% | 21.6% | 44 |
| Germany | 2024 | 18.1% | 58.3% | 23.5% | 50 |
| Brazil | 2025 | 17.2% | 60.3% | 22.6% | 56 |
| India | 2025 | 16.0% | 43.6% | 40.4% | 61 |
| Indonesia | 2023 | 15.8% | 47.8% | 36.4% | 62 |
| Portugal | 2025 | 14.6% | 62.1% | 23.3% | 70 |
| Thailand | 2025 | 13.6% | 51.6% | 34.8% | 74 |
| Czechia | 2024 | 10.8% | 66.8% | 22.3% | 79 |
| Niger | 2022 | 0.8% | 8.5% | 90.7% | 107 |
Income improves the match, and flips what is left
The first half of that sentence is the strongest relationship in the data. The matched share correlates at 0.78 with income per person, with medians of 36.3%, 56.7% and 62.0% from the poorest third of economies to the richest. Richer economies do put more of their workers at their own level, and by a wide margin.
The second half is what the medians show underneath. Under-education falls from 53.6% to 17.5% across those same thirds, while over-education rises from 8.6% to 20.2%, correlating at minus 0.79 and 0.57 with income. The mismatch that remains in a rich economy is the opposite of the one a poor economy carries.
The table holds both shapes at once. Niger and India sit at 90.7% and 40.4% under-educated, while South Korea and Spain sit at 40.9% and 30.1% over-educated, and Poland shows that the two are not simply ordered by income: at 54,262 dollars a head it matches 68.5% of its workers, more than Germany, France or Spain.
What over-education is not
It is not a measure of waste. The comparison is between a worker’s education and the schooling their occupation is defined as needing, and a job can need less schooling than its holder has for reasons that suit both: a first job, a deliberate change of field, a role whose skill sits outside formal education. Nothing here says the worker would be more productive elsewhere.
Nor is it a verdict on the labour market. A low over-educated share can mean jobs that fit, and it can mean an economy whose occupations demand little schooling because there is little schooling to demand. Niger’s 0.8% is the second kind, and the country with the fewest over-educated workers in this set is not the one with the best jobs.
Women are over-educated slightly more often than men, and not everywhere. The women’s share exceeds the men’s in 65 of the 107 economies, with a median gap of 1.8 points. In the table Poland is the widest, 25.6% of employed women against 15.1% of employed men, while in South Korea and the United States the gap runs the other way.
- It compares education with the JOB’S required level, not with the person. Over-educated means the occupation’s normative schooling sits below what the worker holds. It does not mean the worker is too good for the job, and it measures nothing about their skill, their pay or their satisfaction.
- That required level is a convention, applied identically everywhere. The normative approach reads it off the ISCO occupation group, which is what makes 107 countries comparable at all, and it is also what the measure cannot see: a job that genuinely needs more schooling in one country than in another counts the same in both.
- Everyone counted here has a job. This is the fit between education and occupation among the EMPLOYED: it says nothing about how hard that job was to find, and nothing at all about the unemployed, who are not in the denominator.
So Spain’s 30.1% does not say that 30.1% of Spanish workers are wasted, and Czechia’s 10.8% does not say Czech workers are better placed: they say that of every hundred employed people classified in Spain in 2024, 30.1 held more education than their occupation’s normative level, and in Czechia 10.8 did.
Three shares, one denominator, and a reading that forgets what is being compared says what the source does not.
